Small-Intestinal Bleeding

A. General

Upper and lower gastrointestinal bleeding are discussed in Chapters 9 and 12, respectively. Small-bowel lesions are the most common cause of “obscure gastrointestinal bleeding,” defined as hemorrhage that persists or recurs after negative initial upper and lower endoscopies.

B. Diagnosis


Push enteroscopy employs a 400-cm enteroscope to visualize well into the jejunum. Its efficacy is highly dependent on the skill of the endoscopist. It can perform biopsy and therapeutic maneuvers. Intraoperative push endoscopy can sometimes be useful; the scope is introduced through a small enterotomy distal to the ligament of Treitz following exploratory laparotomy.

Extended small-bowel enteroscopy depends on peristalsis to move the scope distally; thus, the procedure may require up to 8 hours for completion. Furthermore, it has no biopsy or therapeutic capabilities. It may visualize as much as 70% of the small intestine and may be more sensitive than conventional enteroclysis.


Capsule endoscopy. In 2001, the Food and Drug Administration approved a disposable “camera pill” that is swallowed and images the entire GI tract as it passes from mouth to anus. A recent meta-analysis suggests that capsule endoscopy is superior to push enteroscopy and barium small-bowel imaging for diagnosing obscure GI bleeding (ACP J Club 2006;144:76).

Double-balloon enteroscopy is a novel method developed in Japan in the late 1990s, It uses two inflatable balloons on the tip of an endoscope to relatively quickly negotiate the small bowel. The technique can be used antegrade or retrograde and allows therapeutic intervention. Although indications and clinical applications are evolving, double-balloon enteroscopy offers the promise that the small bowel may be fully accessible to endoscopic diagnosis and treatment in the same manner as the rest of the GI tract. It also offers the intriguing possibility of endoscopic visualization and/or endoscopic retrograde cholangiopancreatography (ERCP) in segments of the bowel previously inaccessible due to surgery (i.e., the Roux limb and gastric remnant after gastric bypass).


A tagged red-blood cell nuclear medicine scan is highly sensitive for the detection of gastrointestinal bleeding, detecting rates of hemorrhage as low as 0.5 mL/second. However, it is of limited utility in direct resection of a bleeding small-bowel lesion because it lacks anatomic detail.

Angiography has a better ability to localize a bleeding small-bowel lesion, although it is less sensitive than a nuclear medicine study (detects bleeding at 1 mL/second). The angiographer can embolize metal coils into the bleeding mesenteric vessel or leave a catheter in place to assist in intraoperative localization of the lesion. Methylene blue can also be selectively injected to stain the target segment of intestine.

Effective surgical therapy hinges on successful preoperative localization of the bleeding lesion. Unlike the remainder of the GI tract, the small bowel cannot be resected en bloc for intractable bleeding. Preoperative localization of the lesion for segmental resection is strongly advised because it might be very difficult to identify intraoperatively. The angiographic techniques described previously are invaluable in this regard.